Your browser is not supported. Please update it.

2 April 2008

Google puts on brave face but open impact of 700MHz auction looks limited

Google made brave noises about the outcome of the apparently interminable 700MHz auction in the US, but in reality the result was a victory for the status quo and certainly not the dawn of a new era of open access, as promoted by the search/advertising giant. Google’s masterful lobbying and PR antics prior to the auction certainly upped the stakes, probably boosting the net result for the US Treasury coffers, and forcing some open access concessions from the FCC and the victorious major carriers. So far so predictable ‘ as were the gaping holes in those openness plans, as set out by Verizon and AT&T in the immediate aftermath of the auction.

After 38 bidding days and 261 rounds the FCC finally closed the auction of 52MHz of valuable 700MHz spectrum on March 18, and announced the winners last week. 101 different companies spent a total of $19.12bn, substantially exceeding expectations of between $10bn and $15bn. Google, as expected, had dropped out of the bidding for Block C ‘ the one with open access mandates ‘ when it met its reserve price, and Verizon Wireless took the prize. Although we never believed Google actually wanted to own spectrum ‘ its antics were clearly geared to ensuring open access requirements and boosting the value ‘ the victory of Verizon, with its halfhearted approach to openness, will certainly be a setback to the progress of the full open mobile vision, as expounded by Google, in the US.

In the hype surrounding open access prior to the auction, there were genuine hopes that the US might gain a national network that was truly open and would provide a real challenge to the dominance of the big four cellcos. In reality, Google’s ‘big four’ openness demands ‘ open applications and content, open devices, open services (compulsory wholesaling by license holders) and open interconnection ‘ were watered down by the FCC and the conditions attached to Block C now focus mainly on networks being open to any device, not just those tied to the carrier’s network. Even this has fallen short of the PC-style open access vision, since the operators have been busily ‘ and with some justification ‘ claiming that users prefer the ease of use and reliability of a closed device; and that devices and apps can only be allowed on the network once they have been approved for security and reliability purposes. The other blocks of spectrum auctioned do not carry open access mandates anyway.

Against this backdrop, it was hardly surprising that, when the results were finally announced last week, it emerged that Verizon Wireless and AT&T had grabbed the lion’s share of the spectrum. Though both talked up their openness credentials at the same time, Verizon with its first open developers’ conference, neither have joined the Google-driven Open Handset Alliance, which aims to create a standard device platform based on the search giant’s Android software, with a view to decoupling mobile applications, devices and networks (Sprint and T-Mobile have joined the OHA).

Verizon was the biggest winner in the 700MHz auction, gaining the national C Block ‘ 22MHz of spectrum covering 298m people ‘ in addition to a collection of individual market licenses that together cover 171m people. AT&T won more than 150 licenses in the B block of spectrum, which generally covers metropolitan areas. Together, the two carriers bid $16bn, which represents more than 80% of the $19.6bn in total bids and have positioned themselves to widen the gap between themselves and their next challengers, the beleaguered Sprint plus T-Mobile.

Other major investors were satellite TV compay EchoStar, with a well-priced and strategic purchase of most of Block E (see separate item), while Qualcomm spent a substantial amount to enhance its existing 700MHz holdings. Ultimately, the company may have spent too much and did not acquire enough materially to expand its current mobile TV offering, believes investment analyst firm JRPG.

Another interesting knock-on effect is that the two flat rate carriers, Leap and MetroPCS, are likely to resume their on-off merger talks ‘ now that the big four have introduced flat rate charging in a big way, and the two market leaders have enhanced ability to compete in many markets with new capacity, this merger is becoming almost a necessity, despite the hostility between the two suitors. Talks were blocked during the auction under fair competition rules.

While the pre-auction activities of Google, and its pledge to ensure Block C met its reserve price and so triggered the open access conditions, have certainly pushed Verizon and AT&T into openness gestures more quickly than they might have liked, they are still firmly in control of the agenda. Verizon, which has the regulatory obligations on open access, looks likely to do just enough to keep the FCC happy, while AT&T will follow just enough to appear to be playing the mobile internet game ‘ while both operators, like many of their European counterparts, will in practise seek to enhance their closed browsing and content experiences, and work with their pet device makers on highly marketed exclusives, to make it more attractive for the vast majority of users to stay within the walled garden. The success of the AT&T/iPhone tie-up only indicates how right the carrier’s instincts are ‘ that the bulk of the US consumer public is more interested in strong brands and a predictable, pleasant experience than in unfettered access and a free choice of devices. That will change with new patterns of user behaviour, but for now, the mobile world still belongs to the cellcos and their familiar environments.

However, they will need to adapt for the time, in a few years from now, when open access becomes the norm, at least among higher value customers, and they are making the appropriate noises to reassure investors, the FCC and valuable partners, including Google itself. Verizon, although it has not joined the OHA, has said it will support Android devices on its network, and in conjunction with its first open developer conference ‘ which happily, PR-wise, coincided with the announcement of its success in Block C ‘ it released the interoperability specifications for third party devices to connect to its network (and, of course, Android will have to support these in order to be approved on to the Verizon network, like any other design. This is not the openness ‘ or anarchy, as the cellcos perceive it ‘ of the PC or Wi-Fi internet).

However, Verizon is grasping the PR initiative as well as fulfilling its obligations to the FCC. It has now kicked off its Open Development Initiative to let device makers and software houses develop for its network. In making available the specifications for these parties, Verizon said it wanted to see open development devices in customers’ hands by year end, and won’t require consumers who use the new products to sign contracts with the carrier or pay termination fees. The device maker, however, may still choose to require contracts and fees for access to service. Of course, as well as being able to market and present its own closed services and devices more strongly than open access, and fail to approve unwanted applicants, Verizon can also retain a measure of control through pricing, in order to control access to downloadable content. Verizon VCast subscribers can download music and videos for an additional charge of $15 per month and for $25 also get mobile TV. Some premium service packages throw in V CAST at no extra charge. Verizon has not yet provided details on how it will price downloads of data and content under the Open Development Initiative though it is denying it will act as a ‘gatekeeper’. ‘Obviously nothing is free,’ says spokesperson Jeffrey Nelson. ‘We will establish pricing and people who believe they have got products that can make money for them will take advantage of it.’ Some developers fear carriers will restrict access, or raise fees, for content that makes heavy use of the network, and is not within the portal.

Verizon Communications CEO Ivan Seidenberg made the right noises at the first open developer conference, saying: ‘For Verizon to participate fully in the growth of the wireless industry, we need to partner with the inventors and entrepreneurs who are creating these next generation products and services. Our goal is to make our network the on-ramp for the next phase of wireless innovation.’

Detail on the 700MHz auction results:

The best analysis we saw of the results of the auction came from US-based investment research house JRPG, and here we offer a selection of their insights into a process they have monitored very closely.

While we expected Verizon and AT&T to emerge as the major winners and purchasers of spectrum we were surprised that Verizon secured Block C for so little and that AT&T secured significant portions of Block B for so much. Verizon spent the most ($9.4bn) but captured the crown jewel, Block C for a relatively cheap $0.76/pop/MHz. Overall, Verizon bought an average of 36MHz covering a total of 237m pops for $1.10 per MHz pop. AT&T secured enough of Block B licenses (227) to enable it to flush out a nationwide 700MHz-based network, but had to pay far higher per MHz pop prices for a significant number of the largest geographic areas. The company spent a total of $6.64bn. As compared to Verizon, AT&T bought 12MHz covering 176m pops for an average of $3.15 per MHz pop.

Echostar/Dish Network surprised many by acquiring the majority of Block E at an attractive price of $0.55 per MHz pop for a total expenditure of $711m. The purchases position the company to potentially offer a mobile TV product or enhance its attractiveness to AT&T as an acquisition. In our view, it is a clear winner.

Qualcomm secured only a few but strategically important licenses in Block E. However, in doing so, it had to pay an average of $1.36 per MHz pop and spent a total of $558m. The company was also the lone bidder for Block D (the one with public safety mandates, which failed to meet its reserve and will now be reauctioned). We would not view this auction as successful for Qualcomm.

Cox Wireless was the only major cable company to acquire licenses, spending $304m on 22 licenses in Block A and B. MetroPCS effectively got shut out of the auction in all markets except Boston, where it paid a lofty $3.26 per MHz pop. Otherwise the small carriers were unable to secure any spectrum in major markets.

Verizon:
Verizon Wireless began the auction by bidding in multiple areas of Blocks A and B. After the 26th round, the carrier began reapportioning its bidding units, allowing it to begin bidding in the regional portions of Block C during the 29th round. In the 30th round, the company established enough Provisional Winning Bids (PWBs) to enable the cumulative bids for the Block C regions to exceed the previous PWB established in the Nationwide Block C. This collective bid of $4.74bn slightly exceeded the previous PWB established by Google. There were no further bids in either the major regions or nationwide portion of Block C resulting in the award of the block to Verizon. At $0.76 per MHz pop, we view this purchase as a spectrum bargain. Verizon captured 22MHz of nationwide spectrum in the 700MHz spectrum at a price that is well below the estimated $1.06/pop/MHz that AT&T paid for 12MHz in 700MHz in its acquisition of Aloha Partners last fall. Additionally, the average prices paid in Block A and B were $1.16 and 2.67/pop/MHz, respectively. Strategically, through this purchase, Verizon has drawn far closer to parity with AT&T relative to its spectrum holdings and now has more than enough spectrum to create a powerful 4G network. Verizon also spent another $2.6bn on acquiring 32 licenses in Block A. The largest seven licenses (New England, Philadelphia, Washington, DC, Orlando, Miami, Tampa and Atlanta) represented 44% of the dollars spent and had an average price of $1.62 per MHz pop. Finally, the company acquired 67 licenses for $2.1bn in Block B. The three largest purchases (Los Angeles, Chicago, and Miami/Ft. Lauderdale) represented $1.6bn and were purchased at an average price of $4.89/pop/MHz. Chicago alone went for $9.19/pop/MHz.

AT&T:
We had expected that AT&T might be the most active bidder in the auction. We thought it might participate in the bidding in all five of the blocks (not all at once) and would end up acquiring perhaps Block D, E, and substantial parts of Block B. We now believe that AT&T only intended to bid in Block B. Its bidding unit eligibility of 333m was almost half of that purchased by Verizon (590m) and it would have required almost all of its units just to bid for Block C (281m). In reviewing the bidding by round, AT&T started in Block B and never made any bids outside of the block. In the earlier rounds (1-26) it appeared to be bidding against a variety of contenders including Verizon. Because of the limited interference issues surrounding Block B as opposed to those for Block A, we had expected a premium price for B. However, it appears that by having Verizon and many of the other midsize bidders also bidding in B, the prices escalated to the auction high average of $2.67 per MHz pop. Through its clear tenaciousness and financial wherewithal, AT&T eventually won 227 of the 734 licenses in Block B for $6.64bn. It won eight of the top 10 largest licenses (as measured by population) for a total of $2.563bn and an average price/pop/MHz of $4.45. We view these prices as over-inflated and unlikely to justify a good rate of return until well after a new 700MHz-based network is built and up and running (around 2012). At the same time, we believe that AT&T felt that it had to come out of the auction with the 700MHz spectrum needed not only to build the nationwide network but to also hold off the purchases being made by Verizon. Verizon’s early round bidding and the presence of many, albeit smaller, bidders in Block B, had the ultimate effect of driving prices far higher than we believe AT&T expected to pay.

The Verizon open model:

Verizon Wireless’ chief marketing officer Mike Lanman told the open developer conference that the company will have three models in this area:

‘ Retail. This is where a company goes to market with a device and Verizon sets the price of the device. The company also uses Verizon’s existing rate plans.
‘ Wholesale. This is where a company designs their own voice and data pricing plans.
‘ Custom. In this scenario Verizon creates a custom package for a specialized product or service that a company wants to bring to market.

Lanman said that the company understands that by opening up its network to these different models it will see more devices in its test labs and therefore it is expanding its resources and will also use third party certification facilities. He expects the certification process to take about four weeks.